1. Field of the Invention
The present invention relates to a magnetic resonance imaging (MRI) apparatus that takes an internal image of an object using magnetic resonance phenomenon, an image processing apparatus, a computer program product, and a data storing method for the MRI apparatus.
2. Description of the Related Art
Recently, as the development of parallel imaging has proceeded, a magnetic resonance imaging apparatus has become able to take an image of a wide scope at a high resolution in a short time and with a favorable signal-to-noise (SN) ratio. The parallel imaging is a technology of creating an image of a required area by dividing an area of an object into a plurality of areas, simultaneously scanning the divided areas, and synthesizing a plurality of images obtained as a result of the scanning.
The parallel imaging uses a plurality of receiver coils to scan a plurality of areas simultaneously, and the number of receiver coils trends upward to reduce time for scanning. As the receiver coils are increased, the number of channels to be received increases proportionately, and a data volume of k-space data to be collected also increases in proportion to the number of the channels. Consequently, a required capacity of a storage medium (for example, a hard disk) for storing therein the collected k-space data increases day by day.
For this reason, a technology has been proposed for reducing a required capacity of a storage medium for storing therein data by compressing collected data. For example, according to a data compression method disclosed in JP-A 2002-34950 (KOKAI), focusing on a fact that the absolute value of a change volume in individual values of data continuously collected from an object is smaller than the absolute value of a value set in each piece of the data, a difference from reference data to each collected data (difference between the values) is calculated, the bit length is reduced by deleting several upper bits in difference data obtained as a result of the calculation, the bit-length reduced data is stored into a storage medium instead of the actually collected data, accordingly a required capacity of the storage medium is reduced.
However, according to the above conventional technology, a difference from the reference data needs to be calculated with respect to each of collected data piece by piece, which is very inefficient. Therefore, if the technology is applied to a large volume of data, such as k-space data, it may cause a problem that it takes a long time for storing data into a medium.